ZHONG Qi-ding, WANG Dao-bing, LI Guo-hui

(1.China National Research Institute of Food and Fermentation Industries, Beijing 100015, China；2.National Standardization Center of Food&Fermentation Industry, Beijing 100015, China)

?

Rapid Determination of the Stable Oxygen Isotope Ratio of Ethanol in Aqueous Samples

ZHONG Qi-ding1,2, WANG Dao-bing1,2, LI Guo-hui1,2

(1.ChinaNationalResearchInstituteofFoodandFermentationIndustries,Beijing100015,China；2.NationalStandardizationCenterofFood&FermentationIndustry,Beijing100015,China)

During the last years, it has been demonstrated thatδ18O isotope analysis of ethanol in juices and alcoholic beverages provides powerful information to assess their adulteration by addition of water, or to authenticate their geographical origin. However, it is difficult to accurately and conveniently quantify it due to the measurement principle and its water miscibility. To eliminate the impact of water on ethanolδ18O analysis, a porous-polymer-bonded GC column was used to achieve a baseline separation of ethanol and water, then the water was vented by backflush function and the ethanol was converted and analyzed. This is the first manuscript presenting a systematic evaluation and characterization of a method forδ18O isotope analysis of ethanol in aqueous samples by direct injection of the sample (diluted with acetone) into a GC-TC-IRMS system, with the consequent benefits of eliminating sample pre-treatment, minimum sample amounts required and minimum sample throughput. This method has the following advantages: only 70-200 μL of the sample was required for up to 300 possible injections, and high throughput was observed, caused by short analysis time (18 min for each run), the influences of oxygen-containing compounds on ethanolδ18O analysis were eliminated by the use of a capillary-column bonded porous polymer to obtain a baseline separation prior to high-temperature conversion. Precision was determined to be less than 0.5‰ (1σ), and accuracy was validated by spiked samples and proficiency test samples.

ethanol; stable oxygen isotope ratio; aqueous samples; extraction

Stable isotope ratios are useful tools for fighting against the adulteration of food products[1]. In particular, oxygen isotopic proxies are effective for the detection of fruit juices and authentication of alcoholic beverages, because they can provide valuable data on the amount of added water[2-3].δ18O of ethanol is now considered as a reliable internal reference to improve the detection of the watering in wine and fruit juices[4-7]. A major challenge in implementing the use of stable oxygen isotope proxies is the analysis of ethanolδ18O in aqueous samples. In isotope ratio mass spectrometry (IRMS) analysis, ethanol is required to be converted into carbon monoxide (CO) by the unterzaucher reaction, however, in the same reaction conditions, H2O is also converted into CO[8-11]. Considering that H2O can potentially affect the accurate analysis ofδ18O of ethanol in aqueous samples, it is imperative to develop efficient techniques for the extraction and purification of ethanol for the accurate determination of ethanolδ18O.

A distillation procedure for theδ18O analysis of ethanol in alcoholic beverages and fruit juices has been established[12-13]. However, this procedure is time-consuming (more than 4 h) and requires careful laboratory practice to avoid isotopic fractionation during ethanol extraction. The ethanol content in the distillate is merely greater than 95% (volume fraction), so molecular sieves are additional required to trap the co-extracted water, for this reason, an extended time of 24 h is needed. Moreover, although only a few micromoles of pure ethanol is injected for conversion, sample volumes of greater than 350 mL is utilised for distillation.

On the other hand, during recent years, gas chromatography-isotope ratio mass spectrometry (GC-IRMS) coupled with direct sample injection has become an important tool for examining the stable isotopic composition of food ingredients and for assessing the authenticity of food[14-20]. Typically, for ethanolδ13C analysis, capillary GC columns filled with high-polarity stationary phases, such as polyethylene glycol, are employed for separating ethanol from other polar compounds. Ethanolδ13C analysis exhibits advantages of a small sample size, in addition, the pre-extraction of ethanol from other organic compounds is not required. However, a rapid and accurate method for ethanolδ18O analysis by GC-IRMS is still a specialised endeavour to some extent, which is only practiced by a few laboratories worldwide. There are few studies about18O analysis by GC-IRMS reported in the literature, one attempt was reported for the determination ofδ18O of ethanol from aqueous samples by the direct injection mode combined with GC-IRMS, where a DB-FFAP column was used, but the result reported is unfortunately a suspected outlier[6]. Although co-injected water is suggested to not affectδ13C measurements, it is uncertain whether the outlier is attributed to the incomplete separation of water and ethanol[21-22]. Hence, an alternative route of injection without the co-injection of a large amount of water has been established: ethanol in aqueous samples is extracted prior to injection by solid-phase microextraction (SPME). In this case, samples volume greater than 4 mL and extraction time of 60 min are needed[23-24]. However, this methodology has not been widely applied to the analysis of ethanolδ18O in alcoholic beverages and fruit juices. There are several reasons: variation in isotopic fractionation according to the SPME conditions; this methodology is difficult to control[5,21]; the adsorption efficiency of SPME fibres directly depends on the concentration and inversely depends on the aqueous solubility of the analyte, and the dissolved organic species that compete for SPME-active sites will also affects efficiency[24-25]. For these reasons, there is a demand for the development of advanced techniques that can rapidly determine theδ18O of ethanol in aqueous samples from the viewpoints of simplified procedures and marginal sample consumption.

Recently, a rapid method for the determination of waterδ18O in alcoholic beverages by means of GC-IRMS coupled with direct sample injection was reported[26]. On this basis, in order to achieve the online isotope ratio analysis for ethanol without the off-line or SPME pretreatment procedures, this experiment intended to establish a simpler and more rapid method for measuring the oxygen isotope ratios of ethanol in aqueous samples using a porous-polymer-bonded GC column combined with a commercially available gas chromatography-high temperature conversion-isotope ratio mass spectrometry (GC-TC-IRMS).

1　Materials and Methods

1.1Instrumentation

TC/EA-IRMS system: an isotope ratio mass spectrometry (IRMS, Delta V Advantage) coupled with a high-temperature conversion elemental analyzer; Gas Bench II-IRMS system: IRMS connected to a water-CO2equilibration system (Gas Bench II equipment); GC-TC-IRMS system: a Trace GC Ultra system coupled to an IRMS system via a GC-IsoLink and ConFlo IV universal interface, the GC system was equipped with a TriPlus autosampler and a deactivated guard column (2 m×0.25 mm) and a CP-PoraBOND Q column (50 m×0.32 mm×5.0 μm); The GC-Isolink was equipped with a high temperatrue conversion reactor that an alumina (Al2O3) tube (1.5 mm o.d., 320 mm length) packed with Pt and Ni wires, and with auxiliary (‘magic-mix’) gas (1.8% hydrogen in helium, flow rate of 0.5 mL/min) was used; Data were collected using the Isodat 3.0 software. All these above mentioned components were purchased from Thermo Fisher Scientific (Bremen) GmbH (Bremen, Germany), except the CP-PoraBOND Q column from Varian Inc. (Lake Forest, CA, USA).

1.2Reagents and Samples

Absolute ethanol reagent (HPLC grade), referred to as Std-1, was obtained from Sinopharm Chemical Reagent Co., Ltd (Beijing, China). Acetone and methanol (both HPLC grade) were purchased from Duksan Pure Chemicals Co., Ltd (Ansan, South Korea) and Mallinckrodt Baker Inc. (Phillipsburg, NJ, USA), respectively. Deionized water and mineral water were purchased from a supermarket in China.

Three proficiency test samples were obtained from Eurofins Scientific, Nantes (France): dry wine, pure ethanol and plum spirit in 2012 R1, 2012 R2 and 2012 R3, respectively. Edible ethyl alcohol (used as Std-2), Chinese spirit and red wine were purchased from a supermarket in China.

International reference material from the International Atomic Energy Agency (IAEA, Vienna, Austria): VSMOW (water, 0‰), SLAP (water, -55.5‰) and IAEA-601 (benzoic acid, +23.3‰). Molecular sieves (2 mm beads, UOP type 3?) were purchased from Fluka Chemie GmbH, Buchs, Switzerland.

1.3Methods

1.3.1Determination ofδ18O Values for Laboratory Ethanol Working Standards

Std-1 and Std-2 were used as working standards for oxygen isotope ratio analysis, and trace water in the working standards was trapped by storage for at least 24 h on molecular sieves; theδ18O value [calibratedvsVienna Standard Mean Ocean Water-Standard Light Antarctic Precipitation (VSMOW-SLAP)][27]was determined by TC/EA-IRMS system[13].

1.3.2Determination ofδ18O Values for Water

δ18O values of deionized water and mineral water were determined by GasBench Ⅱ-IRMS sysytem. All procedures have been described in the OIV-MA-AS2-12-MOU18 method (2009)[28]. Theδ18OVSMOWvalues of deionized water and mineral water were (-9.85±0.05)‰ and (-19.33±0.12)‰, respectively.

1.3.3Determination ofδ18O Values for Ethanol in Aqueous Samples

δ18O values of ethanol in aqueous samples were determined by GC-TC-IRMS sysytem. The temperature program started at 100 ℃ and was maintained for 2 min, then increased at a rate of 10 ℃/min to 150 ℃ and maintained for 5 min, and then at 20 ℃/min to 200 ℃ and maintained for 2 min. Once the compounds were separated on the GC column and eluted, ethanol was transferred into the conversion furnace of oxygen analysis. The carrier gas helium was set at 1.2 mL/min.

The reactor was set at 1 280 ℃. Elemental carbon was used to provide a reactive layer, which should be well distributed, and this was done after 80 runs by flushing the reactor with high methane gas concentrations.

The ethanol concentration in the working standards and samples were diluted with acetone to give approximately the same peak amplitude (m/z28) between 4 000 and 8 000 mV. For samples containing undissolved compounds, such as wine and a fermented matrix, after dilution, the solutions were filtered using 0.20 μm or 0.45 μm syringe filters[24-25]. Sample solutions (1 μL) were injected (10 μL syringe) in the split mode (1∶20). The injector equipped with a straight-bore inlet sleeve containing a plug of quartz wool (Thermo Scientific, Bremen, Germany) was set to 200 ℃, and the inlet sleeve was cleaned. The quartz wool plug was replaced after 200 injections; meanwhile the GC column was baked at 280 ℃ to remove any retained substances.

The impact of isotopic fractionation caused by injection, chromatography, pyrolysis and gas transfer was minimised following the principle of identical treatment[29-30]. Here, Std-1 was used as a standard and analyzed along with the unknown samples.

1.4Isotopic Standardisation

All results were calculated according to the equationδ18O[‰]=[Rsample/Rstandard-1], whereRis ratio of the heavy-to-light stable isotope in the sample (Rsample) and in the working standard (Rreference). The isotopic values were calculated and normalised against the working standards: Std-1 and Std-2[27].

1.5Data Analysis

Theδ18O values obtained from the GC-TC-IRMS analyses are reported as means (±standard deviation) of triplicate analyses. Statistical analysis was performed usingt-tests to compare two means. A value ofp<0.05 was considered to indicate statistical significance.

2　Results and Discussion

2.1Separation of Major Oxygen-Containing Compounds

For accurate compound specific stable isotope analysis by GC-IRMS, some crucial aspects should be highlighted: 1) suitable ion currents for analysis[20]；2) baseline separation of target analytes with other compounds[15]. Suitable ion currents will help to avoid source-linearity effects caused by dose-dependent isotope fractionation，thus, it is important to adjust the concentration of the analytes in the sample to similar peak heights. As only 6 nmol CO is required for organic oxygen isotopic measurement, and the alcoholic strength is usually higher than 5% in alcoholic beverage samples, it is very convenient to dilute the sample with a common organic solvent—acetone and methanol, which have been used as diluents for the analysis of ethanolδ13C in whisky and brandy samples[31-32]. Owing to its miscibility with water and organic reagents, the baseline separation of ethanol and other oxygen-containing compounds are absolutely required forδ18O analysis, and then the effluents enter a post-column splitter at the end of the GC capillary. From here, the ethanol peak will be diverted to the pyrolysis reactor and converted into CO gas, and other effluents are vented. To check the capacity of the CP-PoraBOND Q column for the separation of ethanol from water and select one proper diluent, equal volumes of water, ethanol, methanol and acetone were mixed. For this test, 0.02 μL mixture containing equal volumes of water, ethanol and acetone was injected into the injection port using a 0.1 μL liquid sampling syringe and then monitored by GC-TC-IRMS. The four oxygen-containing compounds were well separated with a stable background, and the result was shown in Fig.1.

Note: The flat peak is due to monitoring gas injections introduced by the ConFlo Ⅳ interface, and the other peaks are due to the sample gas generated from water, methanol, ethanol and acetone, respectivelyFig.1　Chromatographic of the ion currents at m/z 28, 29, 30

The effluents were observed in the following order by comparison with the retention time of their reagent: water, methanol, ethanol and acetone, with the retention times of 326, 435, 641 and 861 s, respectively. However, the CO peak heights were not similar to each other, albeit with the relative intensities of 1.17∶3.21∶1.21∶1. The peak area signals of these four derived CO peak heights were observed to exhibit the relative peak areas of 1.35∶2.00∶1.10∶1; discrepancy in these values are attributed to the different vapour pressures of these four compounds in the injection port as well as the different column temperatures, while the target ingredient eluted from the GC capillary column. This test aimed to calculate the retention times of these four main oxygen-containing compounds; hence, further studies and detailed data to account for this discrepancy are not presented here.

Both methanol and acetone can be utilised for sample dilution; however, acetone was selected to distribute column load and to successfully eliminate water. Thus, we set the backflush function of GC-TC-IRMS starting at 1 s and 720 s of the run to vent water and acetone, respectively, and switched it off at 550 s to enable the access of ethanol molecules into the pyrolysis furnace. Fig.2 shows the ion chromatogram.

Note: The flat peaks are due to monitoring gas injections introduced by the ConFlo Ⅳ interface; the second and the fifth flat peak serve as the reference points; the other peaks are for control; the chromatogram peak is generated from ethanolFig.2　Chromatographic of the ion currents at m/z 28, 29, 30

Next, the GC column was subsequently baked at 200 ℃ with the aim of eluting higher alcohols, although these oxygen-containing substances can be ignored because of their minute quantity[26]in alcoholic beverages; in particular, the samples had been already diluted with acetone. One of the advantages of this method is that extraction and purification prior to sample injection are not required anymore, which enables the integration of the separation and pyrolysis of ethanol; this integration significantly reduces time and labour consumption. In addition, because of the liquid injection mode and the high efficiency of GC-TC-IRMS, a very small amount of the test portion is adequate for analysis.

2.2Water Removal Efficiency

The carbon reaction is assumed to convert water into CO during ethanol conversion if the baseline separation of ethanol from water cannot be achieved, and the more water contained in the analyte, the more CO will be generated from water; moreover, the more negative theδ18O value of water, the more depleted theδ18O value of ethanol. In fact, the tailing of the water peak on the chromatogram has been reported to affect18O/16O measurements because of the carbon reduction of water[21]; hence, it is imperative to successfully remove water from the target analyte (ethanol) for accurate quantativeδ18O of ethanol. In this study, the water removal efficiency was tested by two approaches: Firstly, Std-1 and deionized water were used to prepare a series of ethanol solutions at different ratios (details shown in Table 1), and then a consistent ethanol concentration of these solutions was achieved by the dilution of acetone before GC-TC-IRMS analysis; Secondly, Std-1 was added into deionized water and mineral water to achieve the same alcohol strength (both of 5%), followed by dilution with acetone as well as the determination of ethanolδ18O.

In the first verification, the effect of water by the carbon reaction on ethanolδ18O analysis was not observed: Table 1 shows the values obtained from theδ18O values of ethanol solutions, the mean value of seven aqueous samples is 18.32‰ (±0.31‰ ), similar to the value measured from pure ethanol (18.51‰), and the maximum difference for the meanδ18O values of ethanol in spiked samples is only 0.37‰, where the water content in the solutions vary from 0 to 99%.

Table 1　δ18O values of ethanol calculated for Std-1 mixed with different percentages of water

Secondly, despite the fact that there is a difference of 9.48‰ in theδ18O value of the water used, the difference of the ethanolδ18O value in the two aqueous samples is only 0.14‰ (results shown in Table 2), and these variations (0.37‰ and 0.14‰) are lower than the measurement precision (1σ).

Table 2　δ18O values of Std-1 calculated for 5% ethanol mixed using two different water samples

All these results demonstrate that the experimental value of ethanolδ18O varies neither with the water content of the samples nor with theδ18O value of water in the samples; therefore, this PoraBond Q column and the relevant GC conditions are concluded to allow for the baseline separation of ethanol with water and acetone, and the effect of water on ethanolδ18O analysis is successfully eliminated. Hence, this method can be applied to analysis ethanolδ18O for any alcoholic beverage sample.

2.3Application to Samples

The precision of the18O/16O measurement for ethanol was determined using edible ethyl alcohol, wine and Chinese spirit samples. Each sample is replicated five times, and good repeatability is obtained (results shown in Table 3).

Table 3　Repeatability and reproducibility of δ18O for several types of ethanol-containing samples

To remove the insoluble matter in wine, syringe filters were used; however, the soluble but non-volatile compounds are also injected with ethanol, which are retained in the GC injection port; hence, a deactivated guard column (retention gap) is utilized. To evaluate the long-term effect on the analytical performance attributed to those non-volatile compounds in the samples, three independent analyses on the same sample were tested at five different days (during this period, more than 150 wine samples were continuously analyzed). Data were obtained under the same instrument conditions with the usage of an identical inlet sleeve and quartz wool, and reproducibility is shown in Table 3 with SD less than 0.5‰, which confirms the method for the ethanolδ18O analysis of spirits and wines.

2.4Method Validation

As international ethanol or alcoholic beverage reference material was not available for ethanolδ18O analysis, it was difficult to validate the described method as done by Werner[8]. The optimal solution was to measure a working standard (as similar as possible to the analyte of interest) and a sample of known isotopic composition for each organic structure to be studied[32]. The standard was mixed with the sample (with knownδ18O values) prior to the injection, and the theoretical and experimental values of ethanol in the mixture were compared[17,33]. Here, Std-1, the red wine sample (Table 3) and their mixtures (ethanol volume from Std-1 were 25%, 50% and 75%, respectively) were prepared and injected into the GC column under identical conditions. Triplicate analyses were performed and were repeated if the SD exceeded 0.5‰. The theoretical values of ethanol (24.19‰, 22.30‰ and 20.41‰) are significantly (p<0.01, Pearson’s correlation test) correlated with those in the spiked samples ((23.71±0.41)‰, (22.01±0.23)‰ and (20.95±0.43)‰, respectively). And as can be seen in Fig.3, theδ18O values of ethanol obtained by this method and the added ethanol proportion in spiked samples are strongly correlated (R2=0.98). Hence, it can be inferred that the developed method is free of any isotopic fractionation, which is suitable for aqueous samples ethanol18O/16O isotope ratio determination.

To further evaluate the accuracy of this method, a number of proficiency tests for oxygen isotopic measurements were performed[34-35]. As shown by the data in Table 4, several laboratory reported ethanolδ13C values exist for each proficiency test sample; however, only a limited number of participants given the ethanolδ18O value; hence, it may be concluded that the procedures for the ethanolδ13C analysis of aqueous samples cannot be directly utilized for ethanolδ18O analysis (except for combustion interface replaced by pyrolysis interface), even with the reported GC/IRMS method[16-18, 20]. The SD (δ18O) value of a non-pure ethanol sample is higher than 2‰, which indicates technical difficulties for ethanolδ18O analysis in aqueous samples.

Fig.3　Correlation of determined δ18O value of ethanol vs the proportation of std-1

Proficiency test samples were analyzed using the developed method, the maximum deviation (less than 0.71‰) between the measuredδ18O value and the mean value from the proficiency test is acceptable according to the tolerance of reproducibility SD (1.0‰) given by a inter-laboratory study[13]. Furthermore,t-test indicated that the results from the presented method are identical within the mean values of the proficiency test (p<0.05).Z-score values demonstrate the validity of the proposed method, which can be used for routine analysis, and is suitable for the authenticity control of juices and alcoholic beverages.

Table 4　δ18O results of the proficiency test samples

Note: a.n:number of laboratories that submitted the ethanolδ18O value;N:number of laboratories that submitted the ethanolδ13C value; b.δ18O value determined by the developed method; c. deviation of the results of a laboratory from the mean value divided by the target standard deviation obtained from an intercomparison test,Z-score =(χ-μ)/σ;Z=0, ideal;Z<2, satisfactorily; 23, results not sufficient

[1]OGRINC N, KOSIR I J, SPANGENBERG J E, et al. The application of NMR and MS methods for detection of adulteration of wine, fruit juices, and olive oil. A review[J]. Analytical & Bioanalytical Chemistry, 2003, 376(4): 424-30.

[2]DUNBAR J, WILSON A T. Oxygen and hydrogen isotopes in fruit and vegetable juices[J]. Plant Physiology, 1983, 72(3): 725-727.

[3]ROβMANN A, RENIERO F, MOUSSA I, et al. Stable oxygen isotope content of water of EU data-bank wines from Italy, France and Germany[J]. European Food Research and Technology, 1999, 208(5/6): 400-407.

[4]CALDERONE G, GUILLOU C. Analysis of isotopic ratios for the detection of illegal watering of beverages[J]. Food Chemistry, 2008, 106(4): 1 399-1 405.

[5]PERINI M, CAMIN F.δ18O of ethanol in wine and spirits for authentication purposes[J]. Journal of Food Science, 2013, 78(6): C839-C844.

[6]CALDERONE G, RENIERO F, GUILLOU C. Isotopic analysis of ethanol: Study on18O/16O measurement using a high-temperature pyrolysis system coupled to an isotope ratio mass spectrometer[J]. Rapid Communications in Mass Spectrometry, 2006, 20(5): 937-940.

[7]JAMIN E, GUéRIN R, RéTIF M, et al. Improved detection of added water in orange juice by simultaneous determination of the18O/16O isotope ratios of water and ethanol derived from sugars[J]. Journal of Agricultural and Food Chemistry, 2003, 51 (18): 5 202-5 206.

[8]WERNER R. The online18O/16O analysis: Development and application[J]. Isotopes in Environmental and Health Studies, 2003, 39(2): 85-104.

[9]SHARP Z D, ATUDOREI V, DURAKIEWICZ T. A rapid method for determination of hydrogen and oxygen isotope ratios from water and hydrous minerals[J]. Chemical Geology, 2001, 178(1): 197-210.

[10]FARQUHAR G D, HENRY B K, STYLES J M. A rapid on-line technique for determination of oxygen isotope composition of nitrogen-containing organic matter and water[J]. Rapid Communications in Mass Spectrometry, 1997, 11(14): 1 554-1 560.

[11]WANG D B, ZHONG Q D, LI G H, et al. A rapid method for determination of stable oxygen isotope ratio of water in alcoholic beverages[J]. Journal of Agricultural & Food Chemistry, 2015. DOI: 10.1021/acs.jafc.5b00636.

[12]CLAUDIA M B, ERIC J, MICHLE L, et al. Study of the influence of alcoholic fermentation and distillation on the18O/16O isotope ratio of ethanol[J]. Journal of Agricultural & Food Chemistry, 2006, 54(2): 279-284.

[13]THOMAS F, JAMIN E, HAMMOND D.18O Internal referencing method to detect water addition in wines and fruit juices: Interlaboratory study[J]. Journal of Aoac International, 2013, 96(3): 615-624.

[14]JUNG J, PUFF B, EBERTS T, et al. Reductive ester cleavage of acyl glycerides GC-C/P-IRMS measurements of glycerol and fatty alcohols[J]. European Food Research and Technology, 2007, 225(2): 191-197.

[15]SESSIONS A L. Isotope ratio detection for gas chromatography[J]. Journal of separation science, 2006, 29(12): 1 946-1 961.

[16]BAUDLER R, ADAM L, ROSSMANN A, et al. Influence of the distillation step on the ratios of stable isotopes of ethanol in cherry brandies[J]. Journal of Agricultural and Food Chemistry, 2006, 54(3): 864-869.

[17]CABAЙERO A I, RECIO J L, RUPéREZ M. Isotope ratio mass spectrometry coupled to liquid and gas chromatography for wine ethanol characterization[J]. Rapid Communications in Mass Spectrometry, 2008, 22(20): 3 111-3 118.

[18]SPITZKE M E, FAUHL-HASSEK C. Determination of the13C/12C ratios of ethanol and higher alcohols in wine by GC-C-IRMS analysis[J]. European Food Research and Technology, 2010, 231(2): 247-257.

[19]SCHIPILLITI L, DUGO P, BONACCORSI I, et al. Authenticity control on lemon essential oils employing gas chromatography-combustion-isotope ratio mass spectrometry (GC-C-IRMS)[J]. Food Chemistry, 2012, 131(4): 1 523-1 530.

[20]AI G, SUN T, DONG X. Gas chromatography/isotope ratio mass spectrometry: Analysis of methanol, ethanol and acetic acid by direct injection of aqueous alcoholic and acetic acid samples[J]. Rapid Communications in Mass Spectrometry, 2014, 28(15): 1 674-1 682.

[21]YAMADA K, YOSHIDA N, CALDERONE G, et al. Determination of hydrogen, carbon and oxygen isotope ratios of ethanol in aqueous solution at millimole levels[J]. Rapid Communications in Mass Spectrometry, 2007, 21(8): 1 431-1 437.

[23]AGUILAR-CISNEROS B O, LPEZ M G, RICHLING E, et al. Tequila authenticity assessment by headspace SPME-HRGC-IRMS analysis of13C/12C and18O/16O ratios of ethanol[J]. Journal of Agricultural and Food Chemistry, 2002, 50(26): 7 520-7 523.

[24]HATTORI R, YAMADA K, HASEGAWA K, et al. An improved method for the measurement of the isotope ratio of ethanol in various samples, including alcoholic and non-alcoholic beverages[J]. Rapid Communications in Mass Spectrometry, 2008, 22(21): 3 410-3 414.

[25]THOMAS B, FREEMAN K H, ARTHUR M A. Intramolecular carbon isotopic analysis of acetic acid by direct injection of aqueous solution[J]. Organic Geochemistry, 2009, 40(2): 195-200.

[26]WANG D B, ZHONG Q D, LI G H, et al. Rapid method for the determination of the stable oxygen isotope ratio of water in alcoholic beverages[J]. Journal of Agricultural and Food Chemistry, 2015, 63(42): 9 357-9 362.

[27]NELSON S T. A simple, practical methodology for routine VSMOW/SLAP normalization of water samples analyzed by continuous flow methods[J]. Rapid Communications in Mass Spectrometry, 2000, 14(12): 1 044-1 046.

[28]OIV. Organisation Internationale de la Vigne et du Vin method number MA-E-AS2-12-MOUO18. (2009). In OIV. (Ed.), Compendium of international methods of analysis of wines and musts e 398, Vol. 2. Paris, France: OIV.available from .

[29]WERNER R A, BRAND W A. Referencing strategies and techniques in stable isotope ratio analysis[J]. Rapid Communications in Mass Spectrometry, 2001, 15(7): 501-519.

[30]PAUL D, SKRZYPEK G, FORIZS I. Normalization of measured stable isotopic compositions to isotope reference scales-a review[J]. Rapid Communications in Mass Spectrometry, 2007, 21(18): 3 006-3 014.

[31]BRAND W A, TEGTMEYER A R, HILKERT A. Compound-specific isotope analysis: Extending toward15N/14N and18O/16O[J]. Organic Geochemistry, 1994, 21(6): 585-594.

[32]BAUDLER R, ADAM L, ROSSMANN A, et al. Influence of the distillation step on the ratios of stable isotopes of ethanol in cherry brandies[J]. Journal of Agricultural and Food Chemistry, 2006, 54(3): 864-869.

[33]CAMIN F, LARCHER R, PERINI M, et al. Characterisation of authentic Italian extra-virgin olive oils by stable isotope ratios of C, O and H and mineral composition[J]. Food Chemistry, 2010, 118(4): 901-909.

[34]CHRISTOPH N, ROSSMANN A, VOERKELIUS S. Possibilities and limitations of wine authentication using stable isotope and meteorological data, data banks and statistical tests. Part 1: Wines from Franconia and Lake Constance 1992 to 2001[J]. Mitt Klosterneuburg, 2003, 53: 23-40.

[35]THOMAS F, JAMIN E.2H NMR and13C-IRMS analyses of acetic acid from vinegar,18O-IRMS analysis of water in vinegar: International collaborative study report[J]. Analytica Chimica Acta, 2009, 649(1): 98-105.

10.7538/zpxb.youxian.2016.0033

快速測定含水溶液中乙醇的氧同位素比值

鐘其頂1,2，王道兵1,2，李國輝1,2

(1.中國食品發(fā)酵工業(yè)研究院，北京100015；2.全國食品發(fā)酵標(biāo)準(zhǔn)中心，北京100015)

近年來，乙醇的氧同位素比值(δ18O)在果汁和飲料酒真實(shí)性鑒別中起著重要作用，利用該指標(biāo)可檢測產(chǎn)品中的外源水、追溯產(chǎn)品的產(chǎn)地。本工作采用多孔聚合物氣相色譜柱實(shí)現(xiàn)了水與乙醇的在線快速分離，建立了溶劑稀釋后直接用氣相色譜-裂解-穩(wěn)定同位素比值質(zhì)譜(GC-TC-IRMS)測定溶液中乙醇δ18O值的方法。實(shí)驗(yàn)結(jié)果表明，該方法可排除水對乙醇δ18O分析的干擾，乙醇濃度在1%～100%時(shí)測定穩(wěn)定性良好，在不同水溶液中乙醇δ18O的測定值保持一致；乙醇δ18O值重復(fù)性和再現(xiàn)性的標(biāo)準(zhǔn)偏差均優(yōu)于0.5‰，并通過歐盟實(shí)驗(yàn)室間能力驗(yàn)證(FIT-PTS)證明了方法的準(zhǔn)確性。該方法具有樣品用量少(僅需70～200 μL)、分析速度快(約18 min)、操作簡單方便等特點(diǎn)，可為乙醇δ18O在果汁和飲料酒真實(shí)性領(lǐng)域的研究與應(yīng)用提供方法參考。

乙醇；穩(wěn)定氧同位素比值；含水溶液；提取

Date：2015-10-15；Accepted Date：2016-01-06

O657.63Document code: AArticle IC: 1004-2997(2016)05-0471-10

Support Projects：Supported by FP7 project food integrity ‘Ensuring the integrity of the European food chain’ (613688) from European Union Author:ZHONG Qi-ding (1980—), male (Han nationality), Guangxi, senior engineer, analysis of stable isotope in food. E-mail: zhongqiding@163.com

Web Publishing Time：2016-07-05;

Web Publishing Address：http:∥www.cnki.net/kcms/detail/11.2979.TH.20160705.1210.010.html